Lindorm:Manage index tables

Key concepts

A collection is an index table in LindormSearch, roughly equivalent to a table in a relational database. Each collection is divided into one or more shards for distributed storage, and each shard can have one or more replicas for fault tolerance.

Prerequisites

Before you begin, ensure that you have:

• Access to the LindormSearch cluster management system

• A Lindorm instance with at least one LindormSearch node

Create a collection

1. Log on to the cluster management system provided by the LindormSearch search engine service.

2. In the left navigation pane, click Collections.

3. Click Add Collection.

1. Configure the collection parameters. The following constraint must hold. If violated, collection creation fails:

   Parameter	Default	Description
   name	—	The name of the collection.
   config set	_indexer_default	The configuration set of the collection.
   numShards	—	The number of shards. Set this to an integer multiple of the number of LindormSearch nodes. For example, with 2 nodes, set numShards to 2 or 4. The value must be less than MaxShardsPerNode.
   replicationFactor	1	The number of replicas per shard.
   maxShardsPerNode	1	The maximum number of shards on each node.
   autoAddReplica	—	Whether to restore shards on other nodes when a node becomes unavailable. Set to True (recommended) to enable automatic shard recovery.

   maxShardsPerNode × Number of nodes ≥ numShards × replicationFactor

2. Click Add Collection in the dialog box.

Write data

1. In the left navigation pane, select the collection from the Collection Selector drop-down list.

2. Click Documents to go to the data writing page.

3. Select CSV from the Document Type drop-down list.

4. Enter data in the Document(s) field. The following example writes three rows:

   id,update_version_l,name_s,age_i
   1,1,zhangsan,10
   2,2,lisi,20
   3,3,wangwu,30

   The id and update_version_l columns are required — they are system-defined columns that LindormSearch uses internally. The suffix _s of name_s indicates that the column is of the String data type, and the suffix _i of age_i indicates that the column is of the Integer data type.

1. Click Submit Document.

Query data

1. In the left navigation pane, click Query.

2. Enter query conditions in the q field.

3. Click Execute Query.

Delete a collection

1. In the left navigation pane, click Collections.

2. Select the target collection and click Delete.

3. In the Collection text box, enter the name of the collection to confirm deletion.

1. Click Delete.